560 NATUEAL HISTORY OF AQUATIC ANIMALS. 



of which is quite drinkiible aod is inhabited by fresh-water fish. Here the Herrings deposit their 

 eggs ill' two or three feet of water; aud they are found, along with the eggs of fresh-water fish, 

 sticking in abundance to such freshwater ]>lauts as Potamogeton. 



"Nature seems thus to offer us a hint as to the way in which a fish like the shad, which is so 

 closely allied to the Herring, has acquired the habit of ascending rivers to deposit its eggs in 

 purely fresh water. 



"If a full female Herring is gently squeezed over a vessel of sea- water, the eggs will rapidly 

 pour out and sink to the bottom, to which they immediately adhere with so much tenacity that, in 

 half an hour, the vessel may be inverted without their dropping out. When spawning takes place 

 naturally, the eggs fall to the bottom and attach themselves in a similar fashion, but at this time 

 the assembled fish dart wildly about, and the water becomes cloudy with the shed fluid of the milt. 

 The eggs become thus fecundated as they fall, and the development of the young within the ova 

 sticking to the bottom commences at once. 



" The first definite and conclusive evidence as to the manner in which herring-spawn is attached 

 and becomes developed that I know of was obtained by Professor AUman and Dr. Macbain in 1862, 

 in the Frith of Forth. By dredging in localities in which spent Herring were observed on the I'st 

 of March, Professor Allman brought up spawn in abundance at a depth of fourteen to twenty-one 

 fathoms. It was deposited on the surface of the stone, shingle, and gravel, and on old shells and 

 coarse shell-sand, and even on the shells of small living crabs and other Crustacea, adhering tena- 

 ciously to whatever it had fallen on. No spawn was found in any other part of the Forth; but it 

 continued To be abundant on both the east and the west sides of the Isle of May up to the 13th of 

 March, at which time the incubation of the ovum was found to be completed in a great portion of 

 the spawn, and the embryos had become free. On the 25tli scarcely a trace of spawn could be 

 detected, and nearly the whole of the adult fish had left the Forth. 



" Professor Allman draws attention to the fact that ' the deposit of spawn, as evidenced by the 

 appearance of spent Herrings, did not take place till about sixty-five days after the appearance of 

 the Herring in the Frith,' and urrives at the conclusion that 'the incubation probably continues 

 during a period of between twenty-five to thirty days,' adding, however, that the estimate must, 

 for the present, be regarded as only approximative. It was on this aud other evidence that we 

 based our conclusion that the eggs of the Herring 'are hatched in at most from two to three weeks 

 after dei)osition.' 



" Within the last few years a clear light has been thrown upon this question by the labors of 

 the West Baltic Fishery Commission, to which I have so often had occasion to refer. It has been 

 found that artificial fecundation is easily practiced, and that the young fish may be kept in 

 aquaria for as long as five months. Thus a great body of accurate information, some of it of a 

 very unexpected character, has been obtained respecting the development of the eggs and the 

 early condition of the young Herring. 



"It turns out that, as is the case with other flsshes, the ])eriod of incubation is closely depend- 

 ent upon warmth. When the water has a temperature of 53° F. the eggs of the Herring hatch 

 in from six to eight days, the average being seven days. And this is a very interesting fact when 

 we bear in mind the conclusion to which the inquiries of the Dutch meteorologists, and, more 

 lately, those of the Scottish Meteorological Society, appear to tend, namely, that the shoals prefer 

 water of about 55° F. At 50° F. the period of incubation is lengthened to eleven days; at 46° 

 F. to fifteen days; at 38° F. it lasts forty days. As the Forth is usually tolerably cool in the 

 month of March, it is probable that Professor Allman's estimate comes very near the truth for the 

 particular case which he investigated. 



